Many of the problems I had with my attempted sled run the other day (see Lots of lessons in a single run) was a result of Orion's enthusiasm to run. He was so excited he couldn't resist trying to 'spin' in his harness, which of course results in a really ugly tangle.
The solution, of course, is to train him to "line out" like a proper sled dog. "Line out" is a cue given to ask a dog to stand, leaning into the harness, to keep the gang-line tight. Orion's behavior the other day proved I didn't focus on that behavior enough while introducing Orion to the harness last year.
I was wishing I had a video camera I could have set up to record today's training session, because it would provide a good example of how easy it is to train a behavior using strictly positive operant conditioning methods - aka "clicker training".
The first step is to "charge the clicker", basically just communicating to the dog that we are starting a training session and there are really cool treats to be earned. Since many of my dogs are already very familiar with clicker training, I charge the clicker by asking the dog to perform very familiar behaviors in exchange for a click and treat. After 3 cued 'sits' followed by 3 cued jumps up onto his dog house I took Orion over to the gang line, and harnesses him up.
Orion actually helped with the process, shoving his head into the neck ring and lifting each foot in turn to properly don the rig. I hooked the harness to the tug-line, then stepped backward a couple of steps. Orion naturally followed my movement and I clicked the moment the tug-line was taught, tossing him a cookie. After two more repetitions he was happily leaning into the gang-line without prompting, so I then hooked up his neck line.
Next, while still standing in front of him, I started working on duration, clicking only when kept the tug line taught. First for 1 second, then for 2, after a few reps waiting for 3, and so on. With duration steadily improving I to a point where he easily held the position for 10 seconds without fidgeting, I started the cue. Since that is a change I reduced the duration criteria to five seconds, and within a few reps he had apparently figured out that the cue asks for the specific behavior.
Next, I started moving around as I gave the cue and he performed the behavior. This is intended to teach him that the behavior should be repeated regardless of what I may be doing. Within a half dozen reps he was giving me a nice 'line out', on cue even when I was standing behind him.
He was doing SO well that I decided to raise the criteria again. Since I was changing one criteria I moved to his side to reduce the pressure, and started shaping a more 'pretty' line-out, clicking and treating only when he was standing with his head pointed straight ahead rather than looking at me. After 6 successful repetitions I ended the session by letting him free run around the kennel, visiting all his kennel mates while I put the harness and gang-line away and moseyed back toward his house.
So, in one training session lasting no more than 20 minutes, he relearned a basic behavior and handily dealt with 4 different changes in criteria. That's fairly remarkable progress.
Of course Orion has some advantages that make this kind of progress possible. He has been clicker trained literally since the day he opened his eyes. He is normally very food motivated, and today even more so I conducted our session about two hours before his supper time.
In the next session I'll start by reducing all the criteria just a little bit, until he remembers what it is that I'm asking. I'll continue to shape that elegant "nose forward" position just because I really like the way it looks (I can be quite vain when it comes to my dogs, and he looks quite noble in profile). I'll also work in increasing duration and latency (the time it takes for the dog to react to the cue and perform the behavior) and probably call it a day.
In future sessions we'll continue to work on duration and latency, and bring in some more changes such as asking him to line out while standing beside another dog, then later while I hook another dog beside him, and even later when I hook up dogs in front or behind him, and so forth.
I have some other dogs who could benefit from similar remedial training, so I think I'll have plenty to do with the dogs over the next few days, even though there isn't enough snow to go mushing.